Group6 - Final Project #2 Visualization

Authors

Affiliation

Ben Hsiao (5133), …

Columbia University

Dr. Scott Spencer

Columbia University

Preliminary setup

Load libraries to access functions we’ll use in this analysis.

library(tidyverse)
library(leaflet)
library(leaflet.providers)
library(plotly)
library(crosstalk)
library(htmltools)

1 Load the datasets

ev_data <- read.csv("ev_registrations.csv", fileEncoding = "UTF-16LE", sep = "\t") # From DOE/BTS
air_data <- read.csv("epa_air_quality.csv")       # From EPA AQS
station_data <- read.csv("charging_stations.csv") # From AFDC

2 Interactive Visualization #2

# --- 1. DATA PREP ---
state_ref <- data.frame(
  Full = toupper(state.name),
  Abb = state.abb,
  Lat = state.center$y,
  Long = state.center$x
)

# Clean Air Data
pm_col_name <- grep("PM2\\.5|PM25", names(air_data), value = TRUE)[1]
air_clean <- air_data %>%
  select(State, PM25 = all_of(pm_col_name)) %>% 
  mutate(PM25 = as.numeric(as.character(PM25)), State_Upper = toupper(str_trim(State))) %>%
  filter(!is.na(PM25)) %>%
  group_by(State_Upper) %>%
  summarise(Avg_Pollution = mean(PM25, na.rm = TRUE)) %>%
  inner_join(state_ref, by = c("State_Upper" = "Full")) %>%
  select(State_Abb = Abb, Avg_Pollution)

# Clean EV Data
ev_clean <- ev_data %>%
  select(State, matches("2023|EV")) %>% rename(EV_Raw = 2) %>%
  mutate(EV_Count = as.numeric(gsub(",", "", EV_Raw)), State_Upper = toupper(str_trim(State))) %>%
  group_by(State_Upper) %>%
  summarise(Total_EVs = sum(EV_Count, na.rm = TRUE)) %>%
  inner_join(state_ref, by = c("State_Upper" = "Full")) %>%
  select(State_Abb = Abb, Total_EVs)

# Clean Station Data
station_agg <- station_data %>%
  mutate(State_Clean = str_trim(State)) %>%
  filter(nchar(State_Clean) == 2) %>%
  group_by(State_Clean) %>%
  summarise(Total_Ports = n()) %>%
  inner_join(state_ref, by = c("State_Clean" = "Abb")) %>%
  select(State_Abb = State_Clean, Total_Ports, Lat, Long)

# --- 2. LOGIC & SORTING ---
master_data <- station_agg %>%
  inner_join(ev_clean, by = "State_Abb") %>%
  inner_join(air_clean, by = "State_Abb") %>%
  mutate(
    # Strategy Logic
    Median_EVs = median(Total_EVs, na.rm = TRUE),
    Is_Priority = case_when(
      Avg_Pollution > 9.0 & Total_EVs < Median_EVs ~ "Top Priority",
      TRUE ~ "Standard Market"
    ),
    Is_Priority = factor(Is_Priority, levels = c("Standard Market", "Top Priority")),
    
    # Hover Label
    Label = paste0("<b>", State_Abb, "</b><br>",
                   "Strategy: ", Is_Priority, "<br>",
                   "Pollution: ", round(Avg_Pollution, 2), "<br>",
                   "Stations: ", Total_Ports)
  ) %>%
  arrange(Is_Priority) 

# --- 3. VISUALS ---

sd <- SharedData$new(master_data, key = ~State_Abb, group = "Tesla_Hybrid")

# CHART COLORS: Strategy (Red/Grey)
chart_colors <- c("Standard Market" = "#999999", "Top Priority" = "#E31937")

# MAP COLORS: Pollution Gradient (Blue -> Yellow -> Red)
# This creates a color palette function based on the Pollution Number
map_pal <- colorNumeric(palette = c("#00FFFF", "#FFFF00", "#FF0000"), domain = master_data$Avg_Pollution)

# A. CHART (Strategy View)
scatter <- plot_ly(sd, x = ~Total_Ports, y = ~Total_EVs,
                   color = ~Is_Priority, colors = chart_colors,
                   marker = list(
                     size = 18, 
                     opacity = 0.9,
                     line = list(
                       color = ~ifelse(Is_Priority == "Top Priority", "#ff0000", "white"),
                       width = ~ifelse(Is_Priority == "Top Priority", 3, 1)
                     )
                   ),
                   text = ~Label, hoverinfo = "text",
                   type = "scatter", mode = "markers") %>%
  layout(
    title = list(text = "<b>Strategic Matrix</b> (Click Red Dots to Locate)", font = list(size = 14)),
    xaxis = list(title = "Infrastructure (Stations)", type = "log"),
    yaxis = list(title = "EV Adoption (Sales)", type = "log"),
    legend = list(orientation = "h", x = 0, y = -0.2),
    plot_bgcolor = "#f5f5f5"
  ) %>%
  highlight(on = "plotly_click", opacityDim = 1, persistent = FALSE)

# B. MAP (Pollution View)
map <- leaflet(sd, width = "100%", height = 500) %>%
  addProviderTiles(providers$CartoDB.DarkMatter) %>%
  addCircleMarkers(
    lng = ~Long, lat = ~Lat,
    # Priority Dots are Bigger (10px), others are small (6px)
    radius = ~ifelse(Is_Priority == "Top Priority", 10, 6),
    
    # COLOR LOGIC: Use the Map Palette (Gradient) based on Pollution
    fillColor = ~map_pal(Avg_Pollution),
    color = "white", # White border for contrast
    
    fillOpacity = 0.9,
    weight = 1,
    popup = ~Label
  ) %>%
  addLegend("bottomright", 
            pal = map_pal, values = ~Avg_Pollution,
            title = "Pollution Level (PM2.5)",
            opacity = 1)

# ASSEMBLE
bscols(
  widths = c(5, 7),
  list(
    div(style="padding-bottom:10px; font-family:Arial; color:#333;", 
        h4("Strategic Filter"),
        p("Left: Priority Markets (Red). Right: Pollution Heatmap.")),
    scatter
  ),
  map
)

Strategic Filter

Left: Priority Markets (Red). Right: Pollution Heatmap.